DE3650192T3 - Device for determining speeds using ultrasound Doppler echography inside a moving fluid - Google Patents

Abstract

A measure window (29-30, FIG. 2d) is provided in a Doppler velocity profile (FIG. 2c) by deleting an initial section (T1) and a terminal section (T2) of each emission cycle of a train of ultrasound of a period T so as to conserve only the section (To) which it is desirable to study. The conserved section (To) is thereafter segmented into a certain number (N) of measure channels. The advantage of restricting the zone divided into channels to the part which is really useful is that the definition of the profile which is obtained is improved, even when the emitter-receptor of ultrasound is located at a relatively important distance in comparison with the depth of the zone which is studied.

Description

The present invention relates to a device for determining the local velocities inside a selected region of a fluid mass in motion containing solid particles by Doppler ultrasonic echo recording, of the type comprising at least one transducer excited during a plurality of cycles to emit periodic ultrasonic wave trains through said fluid along an acoustic axis in successive cycles of period T and to capture the echo of the wave trains reflected by said particles along said axis, and further comprising means for measuring the velocity of the fluid and at least one circuit for displaying on the screen of a cathode ray tube the curve of the velocities of the Doppler effect caused by the velocity of the particles, representative of the velocity of the fluid, as a function of the distance along said axis, the device comprising a generator for a measurement window of given position, length and number of channels, delay means for defining said position of the window, consisting of a delay circuit designed so that in each of the cycles, at least during an initial time period T1 preceding the start of the rising portion of the velocity curve, the transmission at least from the converter to the cathode ray tube of the screen is cut off so that the start of the measurement window is determined by said initial time period T1, the window generator further comprising an oscillator with adjustable frequency for periodically delivering pulses of a predetermined time period τ corresponding to the time resolution of a measurement channel contained in said window, and means for adjusting the frequency of this oscillator, said means for measuring the fluid velocity being arranged so that the velocity is determined for each of the measurement channels contained in the measurement window.

A device of this type is known from the description in the dissertation No. 5746 of the Swiss Federal Institute of Technology in Zurich (1976), which was submitted by P.-E. Doriot and entitled "Blood flow measurement using of a multi-channel ultrasound Doppler device". This device includes a control unit which controls the emission of the ultrasound wave trains and sequentially opens and closes the n measuring channels of the device. The delay and the length of an observation interval are displayed on the screen of a cathode ray tube. In this device, the opening and closing, i.e. the length of the measurement window, are first determined by visual or acoustic means. After the delay time has elapsed, the electronic circuit of this device switches on a digital oscillator whose clock frequency is determined by the length of the window G. The pulses generated in this oscillator serve to divide the window G into 15 intervals of time g = G/15. These intervals are opened and closed sequentially. The length of the window is therefore a passive parameter which is determined by the placement of the first and last channels by means of visual and/or acoustic observation.

We also cite the document: Traité d'Electricité de l'EPFL (Electricity Science of the ETH Lausanne), Volume 5, Analyse et synthèse des systèmes logiques (Analysis and Synthesis of Logical Systems), by Daniel Mange. This document describes the use of FLIP-FLOP and AND elements in quasi-synchronous sequential system circuits.

We also refer to the document: Logic Databook, Volume 1, from National Semiconductor, which describes AND, OR and FLIP-FLOP elements in micro-CMOS technology on pages 3-22, 3-43 and 356.

In an article published in IEEE Transactions on Sonics and Ultrasonics, Volume SU-28, No. 4, July 1981, pages 242 to 244, the authors, A. P. G. Hoeks, R. S. Reneman and P. A. Peronneau, also describe a multi-channel Doppler device with serial data processing. In this device, in order to obtain the best resolution, the width of a measuring channel must correspond to the duration of the ultrasound transmission. In addition, the time at which the first measurement value is recorded can be delayed in increments of 4/3 us compared to the time at which the increments are transmitted.

In the study by F. D. McLeod, published under the title "Multichannel pulse Doppler techniques" on pages 86 to 107 of the book by R. S. Renneman, Cardiovascular Application of Ultrasonics, Proceedings of an International Symposium held at Beerse, May 29 and 30, 1913, North-Holland, Amsterdam 1974, the general principle of a multichannel Doppler technique for medical applications is described, considering the limitations of the method in terms of signal-to-noise ratio, filtering, filter calibration, the effects of stationary walls, etc. Practical measurements are not covered. The channels in this study are not grouped together to form a window that could be positioned in a region of space according to the requirements of the fluid flow.

In the patent specification US 4 324 258 (Huebscher et al.) a device is described with which the flow velocity of a fluid in a vessel can be obtained without error with increased resolution. In this device no window is considered which could be moved in order to restrict the velocity measurement to one or another selected area of the space in which the liquid flows.

The present invention aims to eliminate the above-mentioned disadvantages and to improve the known devices by creating a device that allows very variable observation areas to be examined with optimum resolution in a convenient and reliable manner, but at the same time uses simple and reliable electronics with a very small number of components and therefore has low production costs.

Therefore, the invention is characterized by the features appearing in the sole claim.

These features provide multiple opportunities to actively determine the end of the measurement window and achieve high flexibility in use, while ensuring convenient and fast monitoring of the examination and obtaining examination equipment that is reliable and has low production costs.

Further advantages will become apparent from the description which presents the invention in greater detail below with the aid of drawings which schematically and by way of example illustrate an embodiment of the device.

Fig. 1 is a block diagram of the entirety of this embodiment of the device;

Fig. 2a to 2e are explanatory diagrams;

Fig. 3 is a block diagram of one of the components of the structure of Fig. 1;

Fig. 4a to 4d are explanatory diagrams for the operation of the device according to Fig. 1;

Fig. 5 illustrates an example of use of the device according to Fig. 1;

Fig. 6 shows the diagram of the speeds for the case of Fig. 5; and

Fig. 7 shows schematically the use of a device according to Fig. 1 to determine the vector of velocities at different points of a plane in the fluid mass in motion.

Figure 1 shows this embodiment of the device in block diagram form. At 1 we see a liquid or other fluid in which solid particles are floating and which is in motion as shown by the arrows. This liquid is inside a container, of which a thick wall is visible at 2. At 3 we see schematically the axis of a transducer 4, which serves to generate ultrasonic waves propagated along axis 3 (and obliquely to wall 2) and to capture the echo, i.e. the waves reflected by the particles located on axis 3.

An example of such a transducer is described in the following publication: The design of efficient broad-band piezoelectric transducers; Charles S. Desilets, John D. Fraser, and Gordon S. Kino; IEEE Transactions on Sonics and Ultrasonics, Volume SU-25, No. 3, pages 115 to 125.

The signal of the echo picked up by the transducer 4 controls a series of known devices designed to provide, on the screen 14 of a cathode ray tube, an image representing the Doppler effect due to reflection from the particles in motion, this Doppler effect then being represented by a curve which is also a representation of the velocity component of the fluid along the axis 3 as a function of the distance from the transducer 4 along the axis 3. At the same time, the device supplies, in a known manner, in digital form, the data of this velocity curve to a computer 12 which in turn provides the result of its processing of these data via an output 15, and in particular the absolute velocity of the various portions of fluid measured along the axis 3.

The known means that make it possible to carry out Doppler ultrasound echo recording are represented by elements 7, 8, 9, 10 and 11, which can only be described very briefly since they belong to a known process. Element 7 is an attenuation correction element that serves to compensate for the attenuation of the ultrasound waves along their outward and return paths on axis 3. The attenuation coefficient A is calculated using the following formulas:

A(t) = a(t) · b(t)

a(t) = h · t

b(t) = k · exp (f · t).

In these formulas, t is the time corresponding to each position along axis 3, h is the path length traversed along axis 3, f is the frequency of the ultrasonic vibrations and k is a factor that depends on the nature of the fluid.

Element 8 is a classical demodulator, 9 is an analog-digital converter, and 10 is a digital filter designed to eliminate parasitic frequency components that either originate from certain undesirable noise sources or are caused by movement of the vessel or the pipe in which the liquid is located. An example of such a filter can be found in the following publication: H. Urkoiwitz: Analysis and synthesis of delay line periodic filters, IRE Trans. Circuit Theory, volume CT-4, page 41, 1957. Element 11 is a Doppler shift detector, as used classically in Doppler ultrasound echo recording. Element 12 is a computer.

The device shown further comprises a transmitter 5 which transmits electrical oscillations of frequency f to the converter 4, which converts them into ultrasonic waves. This transmitter simultaneously delivers these oscillations to a device 6, which is shown in detail in Fig. 3.

This device 6 comprises a delay element 20 which is tunable and serves to delay the passage of pulses arriving at 16 by a selected number by suppressing a certain number of the oscillations arriving at it first. An oscillator 22, also tunable, serves to periodically supply time intervals τ. Element 23 is a counter of the pulses received by oscillator 22, this number being the number of observation channels for the Doppler effect with which it is desired to work. This counter 23 is controlled by the delay device 20 in such a way that it only starts counting the desired number of pulses (selected according to the position of its tuning element 26) at the delay produced by 20. This device 6 further comprises a flip-flop 21, an OR gate 24 and an AND gate 25. The operation of this device will be explained below.

When the liquid, as occurs in certain industrial applications, is at a high temperature or is inside a container whose wall is very thick in relation to the layer of liquid to be studied, this presents a problem because in both cases the transducer 4 is located at a distance from the liquid which is very large in relation to the layer of liquid to be studied. liquid layer is considerable, which has an adverse effect on the resolution in Doppler analysis. If the liquid has a high temperature and the wall is relatively thin, a thermally insulating distance must be provided between the wall 2 and the transducer 4.

Fig. 2 shows schematically what happens acoustically in the device according to Fig. 1 and 3. In Fig. 2a, one can see a train 27 of ultrasonic sine waves emitted by the converter 4 and, at 28, the next wave train emitted by the same converter. Here, F is the repetition frequency of the wave trains, T is the repetition interval. In Fig. 2b, one can see the wave trains received back by the converter 4; at 27a there is a first wave train corresponding to 27, at 28a there is the wave train corresponding to 28. These wave trains are slightly longer than shown for 27 and 28 due to internal reflections inside the converter.

Figure 2c shows an example of a graph of the axial velocity components along axis 3 as it appears on the screen of the cathode ray tube 14. The horizontal segment T₁ on the left is the period of time corresponding to zero velocity of the liquid and extending from the beginning of T to the moment when the ultrasonic waves reach the liquid. The rising then falling curve following this first horizontal segment of duration T₁ represents the curve of the velocities along the portion of axis 3 traversing the liquid. It has a duration T₀. A final horizontal segment of duration T₂ corresponds to the time difference between duration T and the time elapsed between the emission of the wave train and the moment when it reaches the end of the liquid region. If the period T were divided into a certain number of measuring channels, as is usual, the resolution would be relatively poor due to the large distance between the transducer and the end of the region under investigation. Since the number of channels is limited, this limitation would have to be accepted. The device 6 has the main purpose of extending the division into a certain number of channels only to the interesting part of the diagram, that is to say to the rising and then falling curve. This is achieved in the following way. The curve (Fig. 2c) is observed in the image screen of the cathode ray tube 14 and causes the device 20 to introduce a delay in the transmission of the pulses picked up by the converter 5 (via 16) which is increased until the beginning of the rising part of the curve in Fig. 2c is reached. This delay is T₁ (Fig. 2c).

On the other hand, in order to eliminate the disturbing influence of the irrelevant portion following the point where the final portion of this curve returns to zero, the oscillator 22 is tuned to deliver pulses of duration τ and the pulse counter 23 is set to the value of N for which the product N · τ gives a time T₀ (Fig. 2c) which corresponds either exactly to the final portion of the curve of Fig. 2c or to a time period as close as possible to this final portion. Let us assume that this is achieved; then the flip-flop 21 (after the delay time T₁) lets pass the pulses coming from the converter 5. Also at this time, the counter 23 starts counting and, as we have seen, this is the time at which the curve of Fig. 2c begins its rising portion. When the curve according to Fig. 2c has reached zero in its final section, the counter 23 has counted the selected number of pulses of duration τ and will now, via its output 26 and the OR gate 24, cause the flip-flop 21 to switch to the OFF position, where it interrupts the passage of the pulses coming to it from 20. If it has not succeeded in obtaining a product τ · N that places the end of the pulse count before the end of the period T (Fig. 2a), the following happens: The beginning of the wave train following the wave train 28, arriving via 16, reaches the OR gate 24 before the counting in 23 is finished, so that now this initial pulse of the ultrasonic wave train causes the flip-flop to switch to the OFF position and interrupt the passage of the pulses coming from 20. In this case, the final part of the count is ineffective because it goes beyond the period T.

Fig. 2d shows schematically the operation of the flip-flop 21, where the time at which it opens the passage is shown at 29 and the time at which it closes it at 30. The time interval T₀ between these two times represents a measurement window. The speed is calculated in the selected N channels during this time T₀. In practice, the number N is chosen to be as high as possible. It is the number N of measurement channels that determines the degree of resolution.

The device 11 supplies at its output in digital form for each of these measurement channels the value of the velocity component along the axis 3, which is firstly sent to the device 13 for digital-to-analog conversion, from where the data is in turn transmitted to the cathode ray tube 14. Secondly, the data supplied by the device 11 are sent in digital form to the computer 12 to be processed there and in particular to calculate the absolute velocity at the measurement points along the acoustic axis 3. An example of such a device is described in the patent US 3,514,747 and in the following publication: Accuracy and limitations of the ultrasonic Doppler blood velocimeter and zero crossing detector; M. J. Lunt; Ultrasound in Medicine and Biology, volume 2, 1975, pages 1 to 10.

Thanks to the window 29, 30, only the curve of the Doppler effect or of the velocity components along the axis 3 appears on the screen of the tube 14 during the time T₀, while the sections with the duration T₁ and T₂, which correspond to the path of the ultrasonic waves outside the liquid, are switched off.

It can be seen that the device described thus makes it possible to concentrate the measuring channels on the useful part of the Doppler curve and to completely ignore what takes place on the path of the ultrasound waves outside the liquid region of interest, thus achieving maximum resolution. This liquid region is not necessarily limited by the vessel wall; it can be more narrowly limited and restricted to a specific liquid region in which it is desired to study in detail a phenomenon related to the velocity distribution. To do this, it is sufficient to set the start 29 and the end 30 of the window to the part of the velocity curve corresponding to this zone.

Fig. 4 shows schematically what happens in practice after the number N has been chosen. In these examples it is assumed that (Fig. 4a) liquid 31 flows in a pipe 32 and a velocity profile according to the curve 33 is present, as it could be determined by the converter 34 using the method just described. Figs. 4b, 4c and 4d correspond to examples of Doppler profiles with different combinations of τ and N values.

The measurement window was chosen as indicated by Tb in Fig. 4b.

For example, initially N = 16 is chosen, which gives us a velocity diagram as shown in Fig. 4b, which fits sufficiently precisely into the measurement window T₀, since 12 channels are allocated to the period T₀ and four channels that would come after are suppressed by the initial pulse of the following ultrasonic wave train, as we saw above. If N = 8 had been chosen, the velocity diagram would have been clipped on the right-hand side and the curve would have been interrupted after the eighth channel, as shown in Fig. 4c, where τ has the same value as in the case of Fig. 4b. On the other hand, if N = 8 and the value of τ' had been chosen to be greater than τ, we would have a diagram as shown in Fig. 4d, where the resolution is less high and two channels are excluded because they come later than the end of Td. Thus, there are three ways to define the end 30 (Fig. 2d) of the window: by choosing N using the counter 23, by choosing τ using the oscillator 22 and by using the beginning of the following ultrasonic wave train.

The frequency f of the ultrasonic waves is determined by the extent of the velocity profile to be measured, by the maximum and minimum sound propagation speed in the media traversed, by the attenuation and by the required resolution. The maximum number Nmax of measuring channels is fixed; it is indirectly related to the Doppler relationship

v = c · fD/2f,

wherein

v = velocity of the fluid

c = speed of sound in the fluid

fD = Doppler frequency

f = used ultrasonic frequency,

and with the Nyquist frequency relation

fD ≤ F/2

where F is the repetition frequency of the ultrasonic wave trains.

It is advantageous to provide the possibility of using two different frequencies, selected according to the conditions: a relatively low frequency to be used for wide measurement profiles and a higher frequency for narrower measurement profiles. This allows a large range of measurement distance and position resolution. This makes it possible to reduce the lowest detectable speed level or speed threshold.

The length and amplitude of the ultrasonic wave trains emitted by the transducer can be adjusted in a known manner according to the damping coefficient A of the fluid (see above) and the required accuracy and resolution.

The number of sinusoids of each ultrasonic wave train can be tuned, as described, for example, in the following publication: F. D. McLeod, M. Anliker: A multiple gate pulse Doppler flowmeter: Proc. IEEE Ultrasonics, Miami, 1971.

The fact of providing a window for the signal picked up, which must be analyzed according to the Doppler echo recording, as described above, has the advantage that there can be a considerable distance between the transducer and the fluid whose velocity profile is to be known, that the area of the fluid to be subjected to Doppler analysis can be very freely can be selected and that with the same number of measuring channels a higher resolution is obtained than in the classic rental method.

Fig. 5 illustrates an example where the device according to Fig. 1 is used in a case where it is desired to know the velocity distribution along several parallel lines in a plane. It is assumed that it is a flow in an angled section of a container or pipe. The various lines 35 represent the trajectories of various points of the liquid mass. The wall of the container or pipe is represented by 36, while at 37 a series of transducers such as 4 (Fig. 1) are represented, arranged in the same plane and with parallel, equidistant axes. The velocity diagram 38 established by Doppler echo recording is shown schematically according to the axes of the transducers 37 in Fig. 6. The velocities in question are of course the components of the real velocity along the acoustic axis of the corresponding transducer.

To produce the diagrams shown in Fig. 6, it is possible to provide either a series of converters 37 and a device as shown in Fig. 1 for each of them, or a series of converters 37 and a single device as shown in Fig. 1, which is used in succession for the various converters thanks to an electronic switch. It is also possible, if the measurement time factor is not of primary importance, to provide a single converter 37 with an electronic device as shown in Fig. 1 and a mechanical control which moves this converter step by step in order to bring it successively into the various positions shown in Fig. 5 for the various converters.

Fig. 7 shows how to determine the vectorial velocity, ie the velocity in its value and direction in a plane, with a group of devices according to Fig. 1. In this figure, at 39, one can see a container angle similar to 36 in Fig. 5. At 40, one can see a series of parallel, equidistant transducers similar to 37 in Fig. 1 and at 41 another series of transducers aligned perpendicular to the direction of the transducers 40. The The measurements carried out by the transducers 40 and 41 make it possible to determine, for each point of intersection of the acoustic axes shown in dashed lines in Fig. 7, the velocity component along each of these axes as well as the resultant. The computer 12 of Fig. 1 makes it possible to calculate these components and their resultant without difficulty.

Of course, one could generalize to the third dimension by arranging a row of equidistant converters above each of the converters 40, 41. For example, if one has five converters for each direction, the result in the three-dimensional case is 5 · 5 · 5 = 125 converters. As explained above, one could work with a smaller number of circles by using point sampling.

Since the measurement time required to obtain the velocity profile along an acoustic axis is very short in relation to the time constant of the flow fluctuations of the fluid (this measurement time is of the order of ten milliseconds), non-stationary and stationary flows can be analyzed equally well.

Claims (1)

1. Device for determining the local velocities inside a selected region of a fluid mass (1) in motion containing solid particles by Doppler ultrasonic echo recording, of the type comprising at least one transducer (4) excited during a plurality of cycles to emit periodic ultrasonic wave trains (27, 28) through said fluid (1) along an acoustic axis (3) in successive cycles of period T and to collect the echo (27a, 28a) of the wave trains reflected by said particles along said axis (3), and further comprising means for measuring the velocity of the fluid and at least one circuit (7 to 11, 13) for displaying on the screen of a cathode ray tube (14) the curve of the velocities of the Doppler effect caused by the velocity of the particles, representative of the velocity of the fluid, as a function of the distance along said axis (3), the device comprising a A generator (6) for a measurement window of given position, length and number of channels, which comprises delay means (20) for determining said position of the window, and which consist of a delay circuit (20) designed such that in each of the cycles, at least during an initial time period T1 preceding the beginning of the rising part of the speed curve, the signal transfer from the converter (4) to the picture tube (14) is cut off in such a way that the beginning of the measurement window is determined by said initial time period T1, the window generator (6) further comprising an oscillator (22) with adjustable frequency to periodically deliver oscillations of a predetermined duration τ corresponding to the time resolution of a measurement channel contained in said window, and means for adjusting the frequency of said oscillator (22), said means for measuring the fluid velocity being arranged in such a way that the velocity is determined for each of the measurement channels contained in the measurement window, characterized in that the window generator comprises an adjustable oscillation counter (23) connected to the oscillator (22) and the delay circuit (20) and means for adjusting this oscillation counter (23) so that the number N of measurement channels can be selected, the end of the measurement window being determined either by the selected duration τ of a channel, by the number N of window channels or by the beginning of the following ultrasonic wave train emitted after a period T if the total time duration T1 + τ · N is longer than the period T, and wherein said window generator (6) comprises an input (16) connected to said converter (4) by means of a transmitter (5) and at least one output connected to said means for Speed measurement (9 to 11), the input (16) of the generator being connected to said delay circuit (20), which in turn is connected to the "ON" input of a flip-flop element (21), the output of which is connected to one of the inputs of an "AND" element (25), the output of which is connected to said means for speed measurement (9 to 11), the output of the oscillator (22) being connected on the one hand to the input of the adjustable oscillation counter (23), which is controlled by the output of said delay circuit (20), and on the other hand to the other input of the "AND" element (25), the output of the oscillation counter (23) being connected to one of the inputs of an "OR" element (24), the other input of which is connected to the input (16) of the window generator (6) and the output of which is connected to the "OFF" input of the flip-flop element (21).